The first phase of the proposed research project will compare an interactive versus a self-supported drug and alcohol (polysubstance) abuse risk assessment in the literature, considering of critical risk indicators reported in the literature. To enhance accuracy, risk information is collected from both the target respondent and a significant other, such as a family member or friend. The two instruments will be psychometrically compared in internal consistency factor structure. Low, moderate and high risk classification profiles will be established for clinical screening. In the second phase of the research, the instrument will be used to screen residents of a large urban community for level of risk. Participants will be randomly assigned within risk groups. The target population adolescents and young adults, males and females aged 15-25, from a demographically and ethnically diverse urban community in New York City. The intervention will be implemented as a video broadcast through a closed-circuit television station serving the community. The video intervention is presented as a puzzle game (called "Face it Together") in which at-risk person and significant other resolve questions and answers about substance abuse together. The videos dramatize families who are either Black, Hispanic or White, and either low or middle SES. The intervention will be broadcast weekly for one month. Outcomes will be evaluated comparing the intervention and control groups after one month and a six months followup. Outcome measures consists of DIS alcohol abuse symptoms, critical substance abuse incidents, MAST, and CAGE scores. The evaluation will be sensitive to qualifications of intervention effects as a function initial risk status and such personal background characteristics as sex, ethnicity, and socioeconomic status.